SIGNIFY: No Ivabradine Effect on Clinical Outcomes in Patients With Stable CAD

August 31, 2014

BARCELONA, SPAIN ( updated ) — Adding the drug ivabradine (Procoralan, Servier) to standard therapy had no effect, overall, on cardiovascular events in a placebo-controlled trial that randomized >19 000 patients with stable CAD[1].

Treatment with the drug was associated with a 10-beat-per-minute drop in heart rate, on average, compared with placebo, as measured after the trial's third month. But over a mean of almost 28 months, 6.8% of ivabradine patients and 6.4% of patients given placebo met the primary end point of cardiovascular death or nonfatal MI (p=0.20). There were no significant outcome differences for the primary end point's components.

"Given that the primary cardiovascular effect of ivabradine is to reduce heart rate, these results suggest that an elevated heart rate is only a marker of risk—but not a modifiable determinant of outcomes—in patients who have stable coronary artery disease without clinical heart failure," according to Prof Kim Fox (Imperial College, Royal Brompton Hospital, London, UK) and colleagues.

Their report was published August 31, 2014 in the New England Journal of Medicine to coincide with Fox's presentation of the trial, called Study Assessing the Morbidity-Mortality Benefits of the If Inhibitor Ivabradine in Patients with Coronary Artery Disease (SIGNIFY), here today at the European Society of Cardiology (ESC) 2014 Congress .

SIGNIFY's "Most Surprising Finding"

Notably, a prespecified SIGNIFY analysis showed ivabradine to be associated with significantly worse outcomes in a subgroup of patients that actually made up more than half the population, those with angina of at least Canadian Cardiovascular Society (CCS) class 2, one of the approved indications for the drug in Europe.

That is the trial's "most surprising finding," according to an editorial accompanying the SIGNIFY paper[2]. "Although this is a subgroup of an overall neutral trial, it is a group of more than 12 000 patients, who were studied where the therapy is approved and in use outside the United States," write Drs E Magnus Ohman and Karen P Alexander (Duke University, Durham, NC).

"Our recommendation is to exercise caution among patients with more severe forms of angina and to consider adjusting beta-blocker doses to effective levels before initiating ivabradine," they write. "The next natural step would be to carry out a second study . . . to ascertain whether this angina subgroup is really one in which we should exercise caution."

Dr Sripal Bangalore (New York University School of Medicine, NY), who isn't connected with the trial, called the hazard observed in the angina group "worrisome, but hypothesis-generating only, given the overall negative results." Given the SIGNIFY overall results, he said in an email to heartwire , "I would say that there is no evidence to support ivabradine use in stable CAD without heart failure."

Panelists discussing the trial after Fox's formal presentation also showed surprise at the subgroup result, even expressing suspicions that it was something of a fluke.

Dr Robert Califf (Duke University Medical Center, Durham, NC) said, "This is sort of a shocking and counterintuitive result that we don't have an explanation for yet." Echoing Ohman and Alexander's editorial, he added, "One has to wonder whether in some way it needs to be replicated. It's very hard to think about replicating a trial that may have harmed, but it seems so out of context with all the rest of the data with this drug."

Fox himself said, "The results are completely counterintuitive. To get a neutral result would be something we wouldn't necessarily be that surprised [about]. But to find that the results go exactly the opposite way to what we had hypothesized is very difficult to try to explain."

Challenge to the "Heart-Rate Hypothesis"

Fox also seemed taken aback by another major implication of the results. "Some years ago there was an editorial in the Lancet that said, the slower [the heart rate], the better. I myself edited a book that said 'slow the heart, beat the disease.' Is that true now?" he asked.

"As always, you need to define the disease. I think there is no doubt that slowing the exercise heart rate is valuable in the management of symptomatic angina. I think there is no doubt that slowing the heart rate in patients with heart failure improves outcomes. What the SIGNIFY study has taught us is that lowering the heart rate in patients with stable coronary disease without heart failure does not prevent the progression of coronary artery disease."

Panelist Dr Thomas Lüscher (University Hospital Zurich, Switzerland), proposed that too much heart-rate reduction may have been the problem. "There may be a limit, and it would be interesting to see the distribution of heart-rate reduction. Maybe those who had an event were in the lower [range]." As can happen with patients who need a pacemaker for bradycardia, a very slow heart rate "may actually endanger coronary flow. That's the only interpretation of these results that comes to my mind."

Hazard in Angina Group Leads to EMA Review

No call for stopping the use of ivabradine in patients like the SIGNIFY subgroup came from Fox, the panelists, or even regulators who had been previously informed of the subgroup analysis findings. In May the European Medicines Agency (EMA) announced it would reassess ivabradine's approval status. Of note, it pointed out that the study allowed a higher dosage of the drug than the 7.5 mg twice-daily maximum dosage approved in Europe. The drug isn't approved for any indication in the US.

Dr Jean-Pierre Bassand , the assigned discussant following Fox's presentation, said the observed hazard in the CCS class >2 group "is a matter of concern" and also proposed that "too high a dose of ivabradine, up to 10 mg [twice daily], may have played a role. But for the time being, we don't know more about the reason that led to this excess of events."

He said the key message in the EMA statement was the one for patients. "While the review is ongoing and pending further communication," he said, quoting the agency, "patients should speak to their doctor or pharmacist if they have any questions or concerns."

83% Also Taking Beta-Blockers

Of the 19 102 patients randomized in SIGNIFY, nearly three-quarters had a history of MI, two-thirds had history of PCI or CABG, and 63.1% entered the study with ""activity-limiting"" angina, at least CCS class 2. The mean LV ejection fraction was 56.4%.

They were assigned to receive ivabradine titrated to a target heart rate of 55 to 60 beats per minute up to 10 mg twice daily or to placebo; ultimately, the mean ivabradine dosage was 8.2 mg twice daily. Beta-blockers were given per guidelines at dosages established for each patient before randomization, as tolerated, and ultimately were taken by 83% of the population.

To enter the trial, patients had to be in sinus rhythm with a heart rate of at least 70 beats per minute and at least one other major CAD risk factor (such as CCS angina class >2 or recent hospitalization for a coronary event) or two minor risk factors (such as low HDL cholesterol, smoking, or age >70 years).

Over a median of 27.8 months, the hazard ratio (HR) for the primary end point was 1.08 (95% CI 0.96–1.20) for the 9550 patients randomized to ivabradine compared with the 9552 given placebo.

The rate of bradycardia was significantly higher in the ivabradine group, 18.0%, vs 2.3% in the placebo group (p<0.001). Of those assigned to ivabradine, 20.6% withdrew from taking the drug (compared with 14.5% who withdrew in the placebo group), primarily because of bradycardia, reported Fox et al.

The primary-end-point rate was 7.6% on ivabradine vs 6.5% for those assigned to placebo among the 12 049 patients who were randomized in CCS class >2 (HR 1.18, 95% CI 1.03–1.35; interaction p=0.02). There was no such significant difference for the other 7053 patients who entered in CCS angina class 1 or had no symptoms. Nor were there significant interactions with outcomes from heart rate, history of MI or revascularization, or whether beta-blockers were on board.

The EMA Review: Unexpected Fallout at the ESC Sessions

The EMA review is at the heart of the ESC's decision not to include the trial at a press conference conducted before its scheduled formal presentation August 31. That press conference did include every other study slated for presentation at the same afternoon session that was to feature SIGNIFY.

The ESC told heartwire in an email that "the SIGNIFY investigators informed the ESC that they were not able to present as part of the official press program . . . while the EMA review is ongoing. As a result, the decision was taken by the ESC to respect the ongoing process and not to include SIGNIFY within press activities."

Of note, the New England Journal of Medicine gave journalists access to the SIGNIFY formal publication two days before the trial's scheduled presentation at the ESC sessions, subject to a strict embargo. Embargoed release of selected reports to the media is routine for many journals.

The study was funded by Servier. "Fox reports personal fees and nonfinancial support from Servier during the conduct of the study; personal fees from AstraZeneca, TaurX, and CellAegis, personal fees and nonfinancial support from Servier and Broadview Ventures, and nonfinancial support from Armgo outside the submitted work. He is also director of Heart Research and Vesalius Trials." Disclosures for the coauthors are listed at . Ohman discloses personal fees from Abiomed, AstraZeneca, Janssen, Pozen, Sanofi, the Medicines Company, and WebMD; and grants and personal fees from Daiichi Sankyo, Eli Lilly, and Gilead Sciences. Alexander discloses receiving grants from Gilead. Bassand discloses being on the speaker's bureau for AstraZeneca.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.